Reading the Canon Like a Lattice: What the Book-by-Book Count Actually Shows

Most claims about what the Bible teaches about God are made by quoting three verses. This is a different method: count them all. The experiment is simple to state. Take every load-bearing text in the canon where the text names what holds everything up, and tag each one. Does it speak of the terminus — “I AM,” first and last, besides me there is none, from everlasting? Does it speak of greatness — “the Lord is great,” unsearchable, mighty? Does it burn the league — “who is like you among the gods,” with the expected answer no one? Does it show a god that gets carried, built, later, upgraded? Does it raise the low and keep them raised? Does it name the non-leaning support — rock, foundation, cornerstone, the one who holds all things?

Six tags. Sixty-six books. The result is not what either the skeptics or the sloppy believers expect, and the interesting part is not the headline. It is the distribution.

What the counts say

Terminus speech is everywhere: it appears in dozens of books, early and late, in law, prophecy, poetry, letters, and apocalypse. The “I AM” grammar is not a few proof-texts; it is the main way the canon talks when it talks about God’s identity. Greatness language is dense too, but it clusters where you would expect a hymnbook — Psalms, the histories — and almost never in the philosophical epistles. The league-burning language appears almost exclusively where idols are physically in the room: Isaiah’s satire, Jeremiah’s parody, Exodus’s song, the Psalms against the nations’ gods. The carried-gods tag is a control group, and it behaves like one: it shows up precisely to be contrasted, the god you carry versus the God who carries.

The two arrows occupy different rooms, and this is the finding that surprised me most when I laid it out. The uplift language — last first, rejected stone, humble exalted, emptied then named above every name — is concentrated in the Jesus tradition, James, 1 Peter, and Philippians 2. It is nearly absent from Torah’s terminus passages. The terminus language never says “I am the smallest chair.” The uplift language never promotes a second floor. If the canon were one author’s consistent theology of “God is the greatest being,” you would expect the superlative to spread evenly. It does not. The grammar splits by function, and the split is exactly what a two-operator structure predicts: a grounding relation and a lifting relation, distinct, both monotone, never confused.

And the silences. A handful of books say nothing in this grammar at all — Song of Songs, Esther, Philemon, the two short letters of John. That is data too. A committee harmonizing a theology would have sprinkled the load-bearing formulas everywhere. A library that carries the same terminus across genres that never needed to mention it, and lets the love poem and the dinner-party note stay silent, looks like what it is: many authors, one subject, no central planning.

What this method is for

The point of counting is not to replace reading with statistics. It is to make a specific claim testable that is usually asserted: that the Bible’s God is uniquely the God who does not lean, and that this is not a proof-text projection. The claim survives being checked. You can run the tags yourself, on your own translation, in an afternoon with a concordance, and try to break the pattern. Find me a terminus text that says God is the biggest of the gods. Find me an uplift text that installs a second floor. Find me a carried-god passage that forgets, for one verse, that the carrying is the indictment. The pattern holds because it is not a theme someone imposed. It is a structure someone found.

This also defuses the tired charge that the whole reading is circular — “you found the pattern because you looked for it.” The tags were chosen for their ordinary meanings, not their conclusions. The distribution was not. Nobody expected the arrows to separate by genre, or the silences to fall where they fall, or the league-burning to correlate with the presence of statues. Those are findings, and they are falsifiable in the good sense: one counterexample per pattern, properly tagged, and the pattern is dead.

The floor, counted

What the count finally shows is the thing the single verses keep saying one at a time. The canon does not rank God. It stops the question. It does this in law courts and lullabies, in siege literature and exile letters, in a Persian palace and a Roman prison, and the consistency across that span is not thematic unity. It is the same terminus, encountered by forty writers who could not have coordinated, described in the only grammar that fits a being who is not on the scale: first, last, I AM, besides me there is none.

The Bible does not argue for the floor. It points at it, sixty-six times, in six different registers, and lets the pattern do the arguing.